Attribution of household viewership information to individuals

Abstract

A system and method for the assignment of person-level viewership. The system receives viewership information describing the viewing of video content at a household. The system additionally receives demographic information for that household, including the numbers of persons associated with the household. For each combination of viewers, the system calculates the probability that the viewers viewed the content based on the demographic attributes of those viewers and the probabilities that individuals sharing those attributes would view the content. The system then attributes the viewing information to one or more persons from the household based on the calculated probabilities. The system additionally updates the probabilities that individuals having different demographic attributes would view the content based on the selection of persons.

We claim: 
     
       1. A computer-implemented method for attributing household viewership information to individuals in a household; the method comprising:
 maintaining, at a computing system, for each of one or more demographic categories a probability that an individual having a possible attribute associated with the demographic category would view a video content, each of one or more demographic categories associated with a plurality of possible attributes; 
 receiving, at the computing system, viewership information describing a viewing event of the video content in a household; 
 receiving, at the computing system, demographic information for the household, the demographic information comprising a number of individuals associated with the household and attributes of each of the individuals; 
 determining, at the computing system, based on the number of individuals associated with the household, possible combinations of viewing groups for the household, wherein each viewing group is comprised of at least one individual associated with the household; 
 calculating, at the computing system, for each of the viewing groups, a probability that the viewing group viewed the video content based on the maintained probabilities for the one or more demographic categories; 
 adjusting, at the computing system, the probabilities for each of the viewing groups by a viewing dependency factor, the viewing dependency factor based on the viewing dependence of the at least one individual in each viewing group; 
 attributing, at the computing system, based on the probabilities for the viewing groups, the at least one individual of the viewing group having the highest probability to the video content; and 
 generating, at the computing system, a report displaying a number of individuals viewing the video content over a period of time. 
 
     
     
       2. The computer-implemented method of  claim 1 , further comprising updating, at the computing system, the maintained probabilities for the one or more demographic categories for the video content based on the attribution of individuals to the video content. 
     
     
       3. The computer-implemented method of  claim 1 , wherein adjusting the probabilities for each of the viewing groups by a viewing dependency factor comprises:
 maintaining, at the computing system, a plurality of Q-factors that express viewing dependence in the household, wherein each Q-factor is associated with the number of individuals associated with the household and a number of viewers of the video content, wherein the number of viewers comprises a number less than or equal to the number of individuals associated with the household; and 
 adjusting, at the computing system, prior to attributing one or more individuals to the video content, the probabilities for each of the viewing groups by the Q-factor associated with a number of viewers corresponding to a number of persons in each viewing group. 
 
     
     
       4. The computer-implemented method of  claim 3 , wherein each Q-factor is further associated with a popularity rating of a show, and wherein adjusting the probabilities for each of the viewing groups comprises selecting a Q-factor based on a popularity rating of the video content. 
     
     
       5. The computer-implemented method of  claim 4 , further comprising adjusting the popularity rating of the video content by a scale factor prior to selecting the Q-factor. 
     
     
       6. The computer-implemented method of  claim 1 , wherein calculating the probability that a viewing group viewed the video content comprises, for each of the one or more demographic categories:
 determining, based on the demographic information, the attributes of each of the individuals associated with the household that correspond to a possible attribute of the plurality of attributes associated with the demographic category; 
 calculating a probability that the at least one individual in the viewing group viewed the video content based on the probabilities for the demographic category and the determined attributes of each individual that correspond to a possible attribute of the plurality of attributes associated with the demographic category; 
 calculating a probability that at least one individual associated with the household viewed the video content based on the probabilities for the demographic category and the determined attributes of each individual that correspond to a possible attribute of the plurality of attributes associated with the demographic category; and 
 calculating, for the demographic category, the probability that the viewing group viewed the video content based on the probability that the at least one individual in the viewing group viewed the video content and the probability that at least one individual associated with the household viewed the video content. 
 
     
     
       7. The computer-implemented method of  claim 6 , wherein the one or more demographic categories comprises at least two demographic categories, and wherein calculating the probability that a viewing group viewed the video content further comprises calculating an average viewing probability for the viewing group based on the probabilities for each of the at least two demographic categories. 
     
     
       8. The computer-implemented method of  claim 1 , wherein the viewing event is a live television broadcast. 
     
     
       9. The computer-implemented method of  claim 1 , wherein the viewing event is a video-on-demand transmission. 
     
     
       10. A non-transitory computer-readable medium encoded with instructions that, when executed by a processor, perform a method for attributing household viewership information to individuals in a household, the method comprising:
 maintaining for each of one or more demographic categories a probability that an individual having a possible attribute associated with the demographic category would view a video content, each of one or more demographic categories associated with a plurality of possible attributes: 
 receiving viewership information describing a viewing event of the video content in a household; 
 receiving demographic information for the household, the demographic information comprising a number of individuals associated with the household and attributes of each of the individuals; 
 determining, based on the number of individuals associated with the household, possible combinations of viewing groups for the household, wherein each viewing group is comprised of at least one individual associated with the household; 
 calculating, for each of the viewing groups, a probability that the viewing group viewed the video content based on the maintained probabilities for the one or more demographic categories; 
 adjusting the probabilities for each of the viewing groups by a viewing dependency factor, the viewing dependency factor based on the viewing dependence of the at least one individual in each viewing group; 
 attributing, based on the probabilities for the viewing groups, the at least one individual of the viewing group having the highest probability to the video content; and 
 generating a report displaying a number of individuals viewing the video content over a period of time. 
 
     
     
       11. The non-transitory computer-readable medium of  claim 10 , wherein the method further comprises: updating the maintained probabilities for the one or more demographic categories for the video content based on the attribution of individuals to the video content. 
     
     
       12. The non-transitory computer-readable medium of  claim 10 , wherein adjusting the probabilities for each of the viewing groups by a viewing dependency factor comprises:
 maintaining a plurality of Q-factors that express viewing dependence in the household, wherein each Q-factor is associated with the number of individuals associated with the household and a number of viewers of the video content, wherein number of viewers comprises a number less than or equal to the number of individuals associated with the household; and 
 adjusting, prior to attributing one or more individuals to the video content, the probabilities for each of the viewing groups by the Q-factor associated with a number of viewers corresponding to a number of persons in each viewing group. 
 
     
     
       13. The non-transitory computer-readable medium of  claim 12 , wherein each Q-factor is further associated with a popularity rating of a show, and wherein adjusting the probabilities for each of the viewing groups comprises selecting a Q-factor based on a popularity rating of the video content. 
     
     
       14. The non-transitory computer-readable medium of  claim 13 , further comprising adjusting the popularity rating of the video content, by a scale factor prior to selecting the Q-factor. 
     
     
       15. The non-transitory computer-readable medium of  claim 10 , wherein calculating the probability that a viewing group viewed the video content comprises, for each of the one or more demographic categories:
 determining, based on the demographic information, the attributes of each of the individuals associated with the household that correspond to a possible attribute of the plurality of attributes associated with the demographic category; 
 calculating a probability that the at least one individual in the viewing group viewed the video content based on the probabilities for the demographic category and the determined attributes of each individual that correspond to a possible attribute of the plurality of attributes associated with the demographic category; 
 calculating a probability that at least one individual associated with the household viewed the video content based on the probabilities for the demographic category and the determined attributes of each individual that correspond to a possible attribute of the plurality of attributes associated with the demographic category; and 
 calculating, for the demographic category, the probability that the viewing group viewed the video content based on the probability that the at least one individual in the viewing group viewed the video content and the probability that at least one individual associated with the household viewed the video content. 
 
     
     
       16. The non-transitory computer-readable medium of  claim 15 , the one or more demographic categories comprises at least two demographic categories, and wherein calculating the probability that a viewing group viewed the video content further comprises calculating average viewing probability for the viewing group based on the probabilities for each of the at least two demographic categories. 
     
     
       17. The non-transitory computer-readable medium of  claim 10 , wherein the viewing event is a live television broadcast. 
     
     
       18. The non-transitory computer-readable medium of  claim 10 , wherein the viewing event is a video-on-demand transmission.